The role of SHIP-1 in T lymphocytes life and death
Geoffrey Gloire1*, Christophe Erneux2 and Jacques Piette1
1
Center for Biomedical Integrated Genoproteomics (CBIG), Virology and Immunology Unit,
University of Liège, 4000 Liège, Belgium
2
Interdisciplinary Research Institute (IRIBHM), University of Brussels, 1070 Brussels,
Belgium.
*Address for correspondence: Geoffrey Gloire, CBIG/GIGA, Virology and Immunology Unit
Institute of Pathology B23, B-4000 Liège Belgium. Email: geoffrey.gloire@ulg.ac.be.
Tel: + 32 4 366 24 26, Fax: + 32 4 366 99 33
Key Words: SHIP-1, T lymphocyte, PI3K, NF-κB, apoptosis, proliferation.
ABBREVIATIONS
PI3K: phosphoinositide 3-kinase; ITIM: immuno-receptor tyrosine inhibition motif; ITAM:
immuno-receptor tyrosine activation motif; Dok: downstream of tyrosine kinase; FcR: IgG Fc
receptor; PLCγ: phospholipase C gamma; BTK: Burton’s tyrosine kinase; GSK-3β: glycogen
synthase kinase 3 beta; KLF2: Krüppel-like factor 2; NF-κB: nuclear factor κB; IκBα :
Inhibitor of κB alpha; TGFβ: transforming growth factor beta.
of neutrophils and monocytes/macrophages
ABSTRACT
SHIP-1, an inositol 5-phosphatase
due to enhanced survival and proliferation
expressed in hemopoietic cells, acts by
of their progenitors.
hydrolysing
contributes to PtdIns(3,4,5)P3 metabolism
the
5-phosphates
from
Although SHIP-1
phosphatidylinositol-3,4,5-triphosphate
in T lymphocytes, its exact role in this cell
(PtdIns(3,4,5)P3)
type is much less explored. Jurkat cells
tetrakisphosphate
and
inositol-1,3,4,5-
(InsP4),
thereby
have recently emerged as an interesting
negatively regulating the PI3K pathway.
tool to study SHIP-1 function in T cells
SHIP-1 plays a major role in inhibiting
because they do not express SHIP-1 at the
proliferation of myeloid cells. As a result,
protein
SHIP-1-/- mice have an increased number
reintroduction experiments in a relatively
level,
thereby
allowing
1
easy to use system. Data obtained from
and
SHIP-1 reintroduction have revealed that
chromosome
SHIP-1 not only acts as a negative player
PtdIns(3,4,5)P3 back to PtdIns(4,5)P2 [5],
in T cell lines proliferation, but also
and
regulate critical pathways, such as NF-κB
phosphatase-1 (SHIP-1), which hydrolyses
activation, and also appear to remarkably
the
inhibit T cell apoptosis. On the other hand,
phosphatidylinositol-3,4,5-triphosphate
experiments using primary T cells from
(PtdIns(3,4,5)P3)
SHIP-1-/- mice have highlighted a new role
tetrakisphosphate (InsP4) [6]. SHIP-1, a
for
cell
145 kDa protein expressed in hemopoietic
development, but also emphasize that this
cells, contains a central phosphatase
protein
domain, an N-terminal SH2 domain and a
SHIP-1
is
in
not
regulatory
required
for
T
T
cell
tensin
by
homologue
10),
deleted
which
SH2-containing
5-phosphates
and
hydrolyses
inositol-5-
from
inositol-1,3,4,5-
proliferation. In support of these results,
C-terminal
SHIP-1-/-
lymphopenic,
phosphorylable NPXY sites and proline-
suggesting that SHIP-1 function in T cells
rich sequences (Figure 1) [7]. SH2 domain
differs from its role in the myeloid lineage.
and NPXY sites mediate the binding of
mice
are
domain
on
containing
SHIP-1 with a series of consensus motifs
or adapter proteins, like immunoreceptor
INTRODUCTION
The PI3K pathway regulates many
signalling motifs (ITIM and ITAM) and
important cellular functions, such as
various Doks [8, 9]. Through its inositol
survival, proliferation, cell activation and
phosphatase activity, SHIP-1 influences
migration [1-3].
Upon activation, PI3K
PtdIns(3,4,5)P3 levels in many cell types,
generates
phosphatidylinositol-3,4,5-
which has a crucial impact in the negative
triphosphate
(PtdIns(3,4,5)P3)
from
regulation of cellular proliferation and
PtdIns(3,4,5)P3 attracts
apoptosis. In support of this, SHIP-1-/-
PtdIns(4,5)P2.
pleckstrin
homology
containing
proteins
(PH)
domain-
mice have an increased number of myeloid
plasma
cells due to enhanced survival and
membrane, which allows their activation
proliferation of their progenitors, thereby
and subsequent signalling.
PH domain-
resulting in splenomegaly and myeloid
containing proteins include notably Akt, a
infiltration in various organs, including
key messenger in cellular proliferation and
lungs [10, 11].
protection from apoptosis [4].
exhibit
to
the
The
SHIP-1-/- myeloid cells
furthermore
decreased
negative regulation of the PI3K pathway is
susceptibility to apoptotic cell death, an
achieved notably by PTEN (phosphatase
observation that has to be linked to the
2
constitutive Akt activation in these cells,
of B cell development.
leading
mechanisms underlying SHIP-1-mediated
to
enhanced
survival
and
The molecular
A
SHIP-1
SH2
Proline-rich
Phosphatase domain
NPXY
NPXY
B
PI3K
PtdIns(4,5)P2
PtdIns(3,4,5)P3
SHIP-1
PtdIns(3,4)P2
PTEN
Akt, Btk, …
Figure 1. A. Structure of the SHIP-1 protein. B. Through its lipid phosphatase activity, SHIP-1
dephosphorylates PtdIns(3,4,5)P3 to PtdIns(3,4,)P2, thereby negatively regulating the PI3K pathway.
PTEN exerts the same function, but catalyzes PtdIns(3,4,5)P3 back into PtdIns(4,5)P2. See text for
details.
proliferation [11].
This observation is
inhibition of B cell proliferation are now
reinforced by the demonstration that a
well understood.
dominant-negative
induced by BCR activation are inhibited by
mutation
reducing
Signalling pathways
catalytic activity of SHIP-1 was found to
co-aggregation
be
FcγRIIB, a low affinity Fc receptor. This
associated
with
acute
myeloid
leukaemia [12].
SHIP-1 also acts at different levels
co-aggregation
containing
between
occurs
immune
BCR
when
complexes
and
IgGbind
in B cell biology. SHIP-1-/- mice exhibit a
simultaneously the BCR and FcγRIIB and
reduction in the number of pre-B and
trigger tyrosine phosphorylation within
immature B cells in the bone marrow, but
immunoreceptor tyrosine-based inhibitory
increased number of hypermature B cells
motifs (ITIM) of the cytoplasmic tail of
in the spleen. B cells from SHIP-1-/- mice
FcγRIIB [17].
are also hypersensitive to BCR stimulation,
induces the recruitment of SHIP-1 to the
but less sensitive to apoptosis [10, 13-16].
tyrosine phosphorylated ITIM of FcγRIIB
These data reflect that the inhibitory
via its SH2 domain.
signals mediated by SHIP-1 have positive
dephosphorylates PtdIns(3,4,5)P3, leading
or negative impacts depending on the stage
to inhibition of PLCγ, Akt and Btk, thereby
This phosphorylation
SHIP-1 then
3
inhibiting
calcium
expression [29-31] has opened up new
mobilisation ([18-20] and [13] for review).
possibilities for studying SHIP-1 biological
functions
FcγRIIB-
role in T cells by complementation
mediated inhibitory signals upon FcεR1-
experiments. Because of their deficiency
induced degranulation in mast cells ([21-
in SHIP-1 expression, Jurkat cells express
23] and [9] for review).
a very high level of PtdIns(3,4,5)P3 and a
SHIP-1
proliferation
also
and
in
subsequent constitutive Akt activation
ROLE OF SHIP-1 IN T
compared to other leukemic cell lines
LYMPHOCYTES
(such as CEM, MOLT-4 or HUT78) or
human PBLs [32]. Restoration of SHIP-1
activity in Jurkat cells was first carried out
Data obtained from T cell lines
Although the role of SHIP-1 has
using a Tet-Off regulated system encoding
been intensely studied in myeloid and B
the lipid phosphatase core of SHIP-1 fused
cells, its exact biological function in T
to the transmembrane region of rat CD2.
cells remains poorly understood. Pioneer
Expression of this membrane-localized
works
tyrosine
SHIP-1 in Jurkat was sufficient to reduce
phosphorylation of SHIP-1 upon CD3 and
Akt constitutive activation and the activity
CD28 stimulation, and redistribution of its
of downstream PI3K effector, constituting
catalytic activity at the plasma membrane
the first evidence that SHIP-1 is capable of
[24, 25], without highlighting a clear
modulating important signalling pathways
biological role of SHIP-1 in the regulation
in T cells [32]. Ectopic expression of the
of T cell activation.
complete
have
demonstrated
a
SHIP-1 was also
SHIP-1
protein
was
also
reported to interact with the ξ chain of the
achieved in Jurkat cells by Horn et al.
TCR, at least in vitro [26]. An embryonic
SHIP-1 expressing Jurkat cells have
work has also demonstrated that FcγRIIB
reduced PtdIns(3,4,5)P3 level compared to
co-aggregation with the TCR results in
the
tyrosine
SHIP-1,
diminution of Akt phosphorylation on
indicating that, like in B cells, SHIP-1
S473 and T308. Phosphorylation of GSK-
would be responsible for FcγRIIB-elicited
3β, an Akt substrate, is also down
inhibitory signalling in T cells [27].
regulated upon SHIP-1 reexpression [33].
Finally, SHIP-1 was demonstrated to play
Interestingly, expression of SHIP-1 in
a role in LFA-1-dependent adhesion [28].
Jurkat also reduces cellular proliferation,
phosphorylation
of
parental
clone,
and
subsequent
The finding that the Jurkat T cell
notably by affecting the G1 cell cycle
line was defective in SHIP-1 (and PTEN)
regulators Rb and p27Kip1 [33] and KLF2, a
4
factor implicated in T cell quiescence [34].
pathway [38]. Importantly, we have also
Altogether,
reintroduction
shown that Jurkat expressing SHIP-1 are
experiments have highlighted a role for
much more resistant to H2O2 and FasL-
SHIP-1 in negatively regulating T cell
induced apoptosis than the parental cells
lines proliferation.
In support of this,
([38] and manuscript under preparation).
down-regulation of SHIP-1 expression has
This result may appear discrepant with
been described in adult T cell leukaemia
previous data obtained with other cell
(ATLL) [35].
types proposing that SHIP-1 increases
these
Using a lentiviral system, we have
apoptosis
by
antagonising
the
anti-
also express SHIP-1 in Jurkat cells and
apoptotic function of PI3K.
have furthermore positioned this protein as
proapoptotic role of SHIP-1 was described
a key adaptor in the activation of NF-κB
in several cell types, including B, myeloid
upon oxidative stress. After exposure to
and erythroid cells [11, 14, 39-41]. On the
H2O2, parental Jurkat cells activate NF-κB
contrary, some authors have reported that
through phosphorylation of the inhibitor
SHIP-1 recruitment attenuates B cell
IκBα on Y42, which triggers its calpain-
apoptosis
mediated degradation and allows the freed
independently of BCR coligation, and that
NF-κB to migrate into the nucleus [36, 37].
failure to recruit SHIP-1 increases the
On the contrary, Jurkat expressing SHIP-1
apoptotic
activate NF-κB through IκB Kinase (IKK)
Furthermore, it is worth noting that SHIP-1
activation and phosphorylation of the
deficient mice are lymphopenic [10, 11,
inhibitor IκBα on S32 and 36 upon H2O2
44], suggesting that, depending on the cell-
stimulation. Cell lines that express SHIP-1
type and the biological context, SHIP-1
naturally (i.e. CEM or the subclone Jurkat
can exert either pro- or anti-apoptotic
JR) also follow the same signalling
functions (Table I).
induced
Indeed, a
by
phenotype
FcγRIIB
[42,
Cell type
Apoptosis inducer
Role of SHIP-1
References
Myeloid cells
Sorbitol,
Pro-apoptotic
[11, 40]
43].
cycloheximide,
anisomycin A, antiFas antibody, TNF-α
Erythroid cells
EPO deprivation
Pro-apoptotic
[39]
B cells
Aggregation of the
Pro-apoptotic
[14]
5
BCR
Activin/TGF-β
Pro-apoptotic
[41]
Clustering of
Anti-apoptotic
[42]
Anti-apoptotic
[38]
FcγRIIB
independently of
BCR coligation
T cells
H2O2
Table I: Pro- or anti-apoptotic role of SHIP-1 in different hemopoietic cells and biological conditions.
number of CD4- and CD8- double negative
cells [44]. On the other hand, there is a
Data obtained from primary T cells
As described above, generation of
reduced number of T cells in the spleen
SHIP-1 deficient mice have considerably
from SHIP-1 KO mice, due to a severe
increased
reduction of CD8+ T cells, suggesting that
our
understanding
of
its
biological role. These mice have notably
SHIP-1
revealed that SHIP-1 does not function
CD4+/CD8+ ratio in the periphery. But the
similarly in the lymphoid than in the
most intriguing feature of CD4+ T cells
myeloid lineage. Indeed, whereas SHIP-1
from SHIP-1 KO mice is that they are
KO mice have more myeloid cells than wt
phenotypically and functionally regulatory
mice, they exhibit a reduction in the T cell
T cells (Tregs) [44].
population [10, 11].
An elegant study
cells from SHIP-1 KO mice are activated,
carried out by the Rothman’s group
but do not respond to TCR stimulation or
recently
addressed
SHIP-1’s
role
in
is
important
in
maintaining
Indeed, splenic T
produce cytokines in response to PMA
primary mouse T cells [44]. These authors
stimulation.
They exhibit furthermore
have demonstrated that, like in B cells,
Tregs markers (such as high TGFβ
SHIP-1 appears to play a biphasic role in T
production
and
expression
of
the
transcription factor Foxp3) and are able to
cell physiology.
First, SHIP-1, together
suppress IL-2 production by CD4+ CD25-
with the adapter protein Dok-1, plays a key
T cells upon CD3 stimulation.
role in T cell development, since mice
from mice deficient for both SHIP-1 and
deficient for both SHIP-1 and Dok-1
Dok-1
exhibit an important reduction in the
highlighting the crucial role of the adaptor
percentage of CD4+ and CD8+ double
Dok-1 for SHIP-1 activity [44].
positive,
analogous cooperation of Dok-1 with
together
with
an
increased
share
the
same
T cells
phenotype,
An
6
SHIP-1 has already been demonstrated in
under certain conditions, in B cells, it
FcγRIIB signalling in B cells, where SHIP-
appears to play an anti-apoptotic role in T
1 serves as an adaptor protein to recruit
cell lines. iii) Moreover, the fact that T
Dok-1 to the receptor, which in turn,
cell specifc loss of PTEN produces some
inhibits the Ras/MAPK pathway [45].
opposite phenotypes to SHIP-1 KO (i.e. T
Altogether, these data prompt us to
cells hyperproliferation) suggests that the
conclude that SHIP-1 functions differently
signalling role of SHIP-1 in T cells is not
in T cells than in myeloid or B cells for at
restricted to its lipid phosphatase activity
least three reasons: i) SHIP-1 does not
with respect to PtdIns(3,4,5)P3,
seem to inhibit T cells proliferation (at
common substrate of the two enzymes
least primary T cells), but instead its
[46].
absence
opposite
creates
an
anergy to
TCR
stimulation. ii) Whereas SHIP-1 is clearly
the
The mechanisms underlying these
roles
investigation
are
in
currently
our
under
laboratory.
a pro-apoptotic protein in myeloid and,
AKNOWLEDGEMENTS
5
This work was supported by grants from
the Belgian National Fund for Scientific
Research (FNRS, Brussels, Belgium), the
Interuniversity Attraction Pole (IAP5/12,
6
Brussels, Belgium), and the ARC program
(ULG 04/09-323, Belgium). GG is a PhD
student from the Télévie program (FNRS,
7
Belgium), EC is supported by the FRIA
(FNRS, Brussels) and JP is Research
8
Director from the FNRS.
REFERENCES
1
2
3
4
Krystal, G. (2000) Semin Immunol
12, 397-403
Ward, S. G. (2006) Trends
Immunol 27, 80-7
Ward, S. G. and Cantrell, D. A.
(2001) Curr Opin Immunol 13,
332-8
Abraham, E. (2005) Crit Care Med
33, S420-2
9
10
Stambolic, V., Suzuki, A., de la
Pompa, J. L., Brothers, G. M.,
Mirtsos, C., Sasaki, T., Ruland, J.,
Penninger, J. M., Siderovski, D. P.
and Mak, T. W. (1998) Cell 95, 2939
Damen, J. E., Liu, L., Rosten, P.,
Humphries, R. K., Jefferson, A. B.,
Majerus, P. W. and Krystal, G.
(1996) Proc Natl Acad Sci U S A
93, 1689-93
Backers, K., Blero, D., Paternotte,
N., Zhang, J. and Erneux, C. (2003)
Adv Enzyme Regul 43, 15-28
Pesesse, X., Backers, K., Moreau,
C., Zhang, J., Blero, D., Paternotte,
N. and Erneux, C. (2006) Adv
Enzyme Regul
Rauh, M. J., Kalesnikoff, J.,
Hughes, M., Sly, L., Lam, V. and
Krystal, G. (2003) Biochem Soc
Trans 31, 286-91
Helgason, C. D., Damen, J. E.,
Rosten, P., Grewal, R., Sorensen,
P., Chappel, S. M., Borowski, A.,
Jirik, F., Krystal, G. and
Humphries, R. K. (1998) Genes
Dev 12, 1610-20
7
11
12
13
14
15
16
17
18
19
20
21
22
Liu, Q., Sasaki, T., Kozieradzki, I.,
Wakeham, A., Itie, A., Dumont, D.
J. and Penninger, J. M. (1999)
Genes Dev 13, 786-91
Luo, J. M., Yoshida, H., Komura,
S., Ohishi, N., Pan, L., Shigeno, K.,
Hanamura, I., Miura, K., Iida, S.,
Ueda, R., Naoe, T., Akao, Y.,
Ohno, R. and Ohnishi, K. (2003)
Leukemia 17, 1-8
Brauweiler, A. M., Tamir, I. and
Cambier, J. C. (2000) Immunol
Rev 176, 69-74
Brauweiler, A., Tamir, I., Dal
Porto, J., Benschop, R. J.,
Helgason, C. D., Humphries, R. K.,
Freed, J. H. and Cambier, J. C.
(2000) J Exp Med 191, 1545-54
Helgason, C. D., Kalberer, C. P.,
Damen, J. E., Chappel, S. M.,
Pineault, N., Krystal, G. and
Humphries, R. K. (2000) J Exp
Med 191, 781-94
Liu, Q., Oliveira-Dos-Santos, A. J.,
Mariathasan, S., Bouchard, D.,
Jones, J., Sarao, R., Kozieradzki, I.,
Ohashi, P. S., Penninger, J. M. and
Dumont, D. J. (1998) J Exp Med
188, 1333-42
Muta, T., Kurosaki, T., Misulovin,
Z., Sanchez, M., Nussenzweig, M.
C. and Ravetch, J. V. (1994) Nature
369, 340
Scharenberg, A. M., El-Hillal, O.,
Fruman, D. A., Beitz, L. O., Li, Z.,
Lin, S., Gout, I., Cantley, L. C.,
Rawlings, D. J. and Kinet, J. P.
(1998) Embo J 17, 1961-72
Aman, M. J., Lamkin, T. D.,
Okada, H., Kurosaki, T. and
Ravichandran, K. S. (1998) J Biol
Chem 273, 33922-8
Bolland, S., Pearse, R. N.,
Kurosaki, T. and Ravetch, J. V.
(1998) Immunity 8, 509-16
Ono, M., Bolland, S., Tempst, P.
and Ravetch, J. V. (1996) Nature
383, 263-6
Huber, M., Helgason, C. D.,
Damen, J. E., Liu, L., Humphries,
23
24
25
26
27
28
29
30
31
32
33
34
35
R. K. and Krystal, G. (1998) Proc
Natl Acad Sci U S A 95, 11330-5
Gimborn, K., Lessmann, E.,
Kuppig, S., Krystal, G. and Huber,
M. (2005) J Immunol 174, 507-16
Lamkin, T. D., Walk, S. F., Liu, L.,
Damen, J. E., Krystal, G. and
Ravichandran, K. S. (1997) J Biol
Chem 272, 10396-401
Edmunds, C., Parry, R. V.,
Burgess, S. J., Reaves, B. and
Ward, S. G. (1999) Eur J Immunol
29, 3507-15
Osborne, M. A., Zenner, G.,
Lubinus, M., Zhang, X., Songyang,
Z., Cantley, L. C., Majerus, P.,
Burn, P. and Kochan, J. P. (1996) J
Biol Chem 271, 29271-8
Jensen, W. A., Marschner, S., Ott,
V. L. and Cambier, J. C. (2001)
Biochem Soc Trans 29, 840-6
Mazerolles, F., Barbat, C., Trucy,
M., Kolanus, W. and Fischer, A.
(2002) J Biol Chem 277, 1276-83
Bruyns, C., Pesesse, X., Moreau,
C., Blero, D. and Erneux, C. (1999)
Biol Chem 380, 969-74
Sakai, A., Thieblemont, C.,
Wellmann, A., Jaffe, E. S. and
Raffeld, M. (1998) Blood 92, 34105
Astoul, E., Edmunds, C., Cantrell,
D. A. and Ward, S. G. (2001)
Trends Immunol 22, 490-6
Freeburn, R. W., Wright, K. L.,
Burgess, S. J., Astoul, E., Cantrell,
D. A. and Ward, S. G. (2002) J
Immunol 169, 5441-50
Horn, S., Endl, E., Fehse, B.,
Weck, M. M., Mayr, G. W. and
Jucker, M. (2004) Leukemia 18,
1839-49
Garcia-Palma, L., Horn, S., Haag,
F., Diessenbacher, P., Streichert,
T., Mayr, G. W. and Jucker, M.
(2005) Br J Haematol 131, 628-31
Fukuda, R., Hayashi, A.,
Utsunomiya, A., Nukada, Y.,
Fukui, R., Itoh, K., Tezuka, K.,
Ohashi, K., Mizuno, K., Sakamoto,
8
36
37
38
39
40
41
M., Hamanoue, M. and Tsuji, T.
(2005) Proc Natl Acad Sci U S A
102, 15213-8
Takada, Y., Mukhopadhyay, A.,
Kundu, G. C., Mahabeleshwar, G.
H., Singh, S. and Aggarwal, B. B.
(2003) J. Biol. Chem. 278, 2423324241
Schoonbroodt, S., Ferreira, V.,
Best-Belpomme, M., Boelaert, J.
R., Legrand-Poels, S., Korner, M.
and Piette, J. (2000) J Immunol
164, 4292-300
Gloire, G., Charlier, E., Rahmouni,
S., Volanti, C., Chariot, A., Erneux,
C. and Piette, J. (2006) Oncogene
Boer, A. K., Drayer, A. L. and
Vellenga, E. (2001) Leukemia 15,
1750-7
Gardai, S., Whitlock, B. B.,
Helgason, C., Ambruso, D., Fadok,
V., Bratton, D. and Henson, P. M.
(2002) J Biol Chem 277, 5236-46
Valderrama-Carvajal, H.,
Cocolakis, E., Lacerte, A., Lee, E.
H., Krystal, G., Ali, S. and Lebrun,
J. J. (2002) Nat Cell Biol 4, 963-9
42
43
44
45
46
Pearse, R. N., Kawabe, T., Bolland,
S., Guinamard, R., Kurosaki, T.
and Ravetch, J. V. (1999)
Immunity 10, 753-60
Ono, M., Okada, H., Bolland, S.,
Yanagi, S., Kurosaki, T. and
Ravetch, J. V. (1997) Cell 90, 293301
Kashiwada, M., Cattoretti, G.,
McKeag, L., Rouse, T., Showalter,
B. M., Al-Alem, U., Niki, M.,
Pandolfi, P. P., Field, E. H. and
Rothman, P. B. (2006) J Immunol
176, 3958-65
Tamir, I., Stolpa, J. C., Helgason,
C. D., Nakamura, K., Bruhns, P.,
Daeron, M. and Cambier, J. C.
(2000) Immunity 12, 347-58
Suzuki, A., Yamaguchi, M. T.,
Ohteki, T., Sasaki, T., Kaisho, T.,
Kimura, Y., Yoshida, R.,
Wakeham, A., Higuchi, T.,
Fukumoto, M., Tsubata, T., Ohashi,
P. S., Koyasu, S., Penninger, J. M.,
Nakano, T. and Mak, T. W. (2001)
Immunity 14, 523-34
9